Differential Form · aiwiki.page

A differential form is a mathematical object that can be integrated over an oriented curve, surface, or higher-dimensional space. On a smooth manifold, a differential form of degree $k$ assigns to each point an alternating multilinear function of $k$ tangent vectors, varying smoothly with the point. Forms provide a coordinate-independent language for calculus, linking differentiation and integration with differential geometry and topology. Their central operations are the wedge product, exterior derivative, and pullback. (arxiv.org)

Definition and local expression

Let $M$ be a smooth manifold of dimension $n$ . At a point $p$ , its tangent space $T_pM$ is a vector space of infinitesimal directions. The dual space $T_p^*M$ , called the cotangent space, consists of linear functions on these directions. A $k$ -form at $p$ is an alternating multilinear map

\omega_p:(T_pM)^k\longrightarrow\mathbb R.

Alternating means that exchanging two arguments reverses the sign; consequently, the value vanishes whenever two arguments coincide. A smooth differential form is thus a particular kind of covariant tensor field. (arxiv.org)

In coordinates $x^1,\ldots,x^n$ , every $k$ -form has a unique expression

\omega=\sum_{i_1<\cdots<i_k} a_{i_1\cdots i_k}(x)\, dx^{i_1}\wedge\cdots\wedge dx^{i_k},

with smooth coefficient functions. The coordinate differentials form a basis of each cotangent space. At a point, the space of $k$ -forms has dimension $\binom nk$ , and forms of degree greater than $n$ vanish. The notation $\Omega^k(M)$ denotes the space of smooth $k$ -forms. A $0$ -form is simply a smooth real-valued function. (ocw.mit.edu)

Wedge product

The exterior algebra of covectors supplies the wedge product, denoted $\wedge$ . It combines a $k$ -form $\alpha$ and an $\ell$ -form $\beta$ into a $(k+\ell)$ -form and satisfies

\alpha\wedge\beta=(-1)^{k\ell}\beta\wedge\alpha.

It is associative and distributive. For coordinate $1$ -forms,

dx\wedge dy=-dy\wedge dx,\qquad dx\wedge dx=0.

On a plane,

(a\,dx+b\,dy)\wedge(c\,dx+e\,dy) =(ae-bc)\,dx\wedge dy.

The coefficient is a determinant, reflecting the signed-area interpretation of an alternating $2$ -form. This antisymmetry distinguishes wedge multiplication from ordinary multiplication and the unrestricted tensor product. (math.stanford.edu)

Exterior derivative

The exterior derivative is a linear operator

d:\Omega^k(M)\longrightarrow\Omega^{k+1}(M).

On functions it gives

df=\sum_i\frac{\partial f}{\partial x^i}\,dx^i,

where the coefficients are partial derivatives. For a coordinate expression $\omega=\sum_I a_I\,dx^I$ , it is defined by

d\omega=\sum_I da_I\wedge dx^I.

Although this formula uses coordinates, the resulting operator is intrinsic. It obeys the graded product rule

d(\alpha\wedge\beta) =d\alpha\wedge\beta+(-1)^k\alpha\wedge d\beta

and the fundamental identity $d^2=0$ . (ocw.mit.edu)

For example,

d(P\,dx+Q\,dy) =\left(\frac{\partial Q}{\partial x} -\frac{\partial P}{\partial y}\right)dx\wedge dy.

This resembles planar curl. In three-dimensional Euclidean space, appropriate identifications of forms with scalar and vector fields express gradient, curl, and divergence through the same operator $d$ . Unlike these identifications, exterior differentiation itself needs neither a metric tensor nor an orientation. (math.stanford.edu)

Pullback and integration

A smooth map $F:M\to N$ transports forms on $N$ back to $M$ through the pullback:

(F^*\omega)_p(v_1,\ldots,v_k) =\omega_{F(p)}(dF_pv_1,\ldots,dF_pv_k).

Here $dF_p$ is the linear map induced on tangent spaces. Pullback preserves wedge products and commutes with exterior differentiation:

F^*(d\omega)=d(F^*\omega).

It exists even when $F$ is not invertible or the manifolds have different dimensions. (ocw.mit.edu)

A $k$ -form is integrated over an oriented $k$ -dimensional domain by pulling it back into coordinate parametrizations. For a curve $\gamma:[a,b]\to M$ , this gives the line integral

\int_\gamma\omega =\int_a^b\omega_{\gamma(t)}(\gamma'(t))\,dt.

Changing coordinates introduces the signed determinant of the Jacobian matrix. Reversing orientation reverses the integral’s sign. Integration of forms therefore differs from integration against a positive measure, which uses absolute Jacobian determinants. (arxiv.org)

The generalized Stokes theorem states that, for a compact oriented $k$ -manifold $S$ with boundary and a smooth $(k-1)$ -form $\eta$ ,

\int_S d\eta=\int_{\partial S}\eta,

where the boundary carries its induced orientation. It includes the fundamental theorem of calculus, Green’s theorem, the classical surface Stokes theorem, and the divergence theorem. (math.mit.edu)

Closed forms and global topology

A form is closed if $d\omega=0$ , and exact if $\omega=d\eta$ . Every exact form is closed because $d^2=0$ . The Poincaré lemma says that closed forms of positive degree are locally exact; they are also exact on star-shaped open subsets of Euclidean space. Global exactness, however, can fail. (math.mit.edu)

On the punctured plane,

\omega=\frac{-y\,dx+x\,dy}{x^2+y^2}

is closed, but its integral around the counterclockwise unit circle is $2\pi$ . It cannot be the differential of a globally defined real-valued function, because such a differential integrates to zero around every closed curve. Locally it is the differential of an angular coordinate. (math.mit.edu)

De Rham cohomology records this obstruction through the quotient vector space

H_{\mathrm{dR}}^k(M) =\frac{\ker(d:\Omega^k\to\Omega^{k+1})} {\operatorname{im}(d:\Omega^{k-1}\to\Omega^k)}.

Two closed forms represent the same class when their difference is exact. Thus differential operations on smooth forms encode global topological information that cannot be detected in a single coordinate neighborhood. (math.mit.edu)

References

Lecture Notes on Differential Formsarxiv.org
Differential Forms and Hodge Theorymath.mit.edu
Thomas Church - Notesmath.stanford.edu
Lecture Notes — Analysis IIocw.mit.edu
02 lecture notes, Fall 2021math.mit.edu